The Truth Hidden in Sheikh Jamal's Shot Map: A Match Won on the Board but Lost in the Numbers

The day is still vivid in my memory from 2026. Chattogram Abahani beat Sheikh Jamal Dhanmondi 2-1 at their home ground. The scoreboard said Abahani won. But when I manually logged all 14 shots from that match and built the first table for xG Chattogram, the numbers told a different story. Abahani scored 2 goals from 1.3 xG, while Sheikh Jamal generated 1.9 xG from 11 shots — meaning Sheikh Jamal had the better goal-scoring probability but still lost. That post was shared 5,200 times and received 1,100 comments. Realizing that new media valued verifiable numbers over hot takes, I began treating every local match as a dataset.

In this article, I will return to that old match's data — because the same pattern repeats in Bangladeshi domestic football and cricket. The official result tells one story, the data tells another. Which is truer is hard to say, but at least we can identify where each story is weak.

Context: What the scoreboard says vs what the metrics say

In 2026, I was a statistics student at Chattogram University. I had just started the Facebook page xG Chattogram. The Sheikh Jamal vs Chattogram Abahani match was an ordinary day in the local league, but for me it was the first chance to hand-log every shot of an entire match.

Simply put, xG models want to show the probability of a shot becoming a goal before it is taken. Distance, angle, shot type (header, boot, volley), whether there is pressure — these variables are factored in. My early model was simple, manually assigning a value to each shot. But that was not the point. The point was that someone had created a verifiable table for this league's match for the first time, which others could check.

| Team | Goals | Shots | xG | xG per shot | |-----|-----|-----|-----|----------------| | Chattogram Abahani | 2 | 3 | 1.3 | 0.43 | | Sheikh Jamal Dhanmondi | 1 | 11 | 1.9 | 0.17 |

Here is the first conflict. Abahani's xG per shot is much higher — meaning they shot from more dangerous positions with fewer shots. Sheikh Jamal took more shots but most from low-probability positions. This is the classic error — measuring dominance by attack volume.

Core analysis: What lies behind 1.9 xG yielding 1 goal

I still return to this match's data because it was my first big lesson — xG does not just explain outcomes, it also serves as a warning.

Sheikh Jamal's 1.9 xG yielding 1 goal means they wasted 0.9 xG. There are three plausible explanations, and each requires separate data.

1. Finishing weakness: The striker missed chances. To detect this, we need goalkeeper save data, which I did not have. But shot locations show how many shots went inside the frame.

2. Poor shot selection: Inflating xG with many low-quality shots. This is the most probable explanation. 11 shots yielding only 1.9 xG means an average of 0.17 per shot — one goal probability every 6 shots. This is behavioral evidence that they shot frequently from distance or bad angles.

3. Defensive coaching: Abahani's defense forced shot selection, creating conditions for shots from low-quality positions under pressure. Detecting this requires pressing metrics.

My model was not advanced enough then to separate the three causes. But one thing was clear — Sheikh Jamal's loss was not bad luck, it was a repeatable pattern. They took too many weak shots and missed a few good chances.

Another aspect many overlook — the timeline of golden chances. If I recall the goal minutes, Abahani's second goal came late in the match, when Sheikh Jamal was pushing forward and leaving space behind. This is not directly captured in xG but important as context.

Contrarian angle: Correlation and causation are different things

Here I must go against myself. There is a correlation between xG deficit and losing — but correlation does not equal causation.

A 1.9 xG yielding 1 goal in a single match could be misfortune. Suppose a striker hit the post, a defender cleared off the line, a referee made a wrong penalty decision — if these happen repeatedly, xG deficit grows but the fault may not lie with the attacker.

I cannot draw permanent conclusions from one match's data. If Sheikh Jamal scores 1.8 goals from 1.9 xG in the next 10 matches, then the 2026 deficit was an outlier. But if they show the same deficit in the next 10 matches, then it is a systemic problem.

Finding such longitudinal data in Bangladeshi domestic leagues is nearly impossible. Because nobody manually logs shots across a season. I did a few matches in 2026-18, but could not sustain it across a full season — because it was not sustainable for one person. That is the real problem. Building a story from one match's data is easy, but building a system requires continuity, funding, and a team.

Another blind spot — my 2026 model was hand-made, simple, and had no venue adjustment. Pitch condition, weather, night match vs day match — these variables were not captured. Playing on Chattogram's rain-soaked pitch is a different context from a dry pitch. My model could not capture that. So I had to be cautious with that table.

Takeaway: What to watch in the next round

What the Sheikh Jamal vs Abahani match data taught me is: goals tell the outcome, xG tells the process, and xG per shot tells the strategy. Separating the three layers reveals the match's truth.

Next time you watch a Bangladeshi domestic match, I will ask readers — are you only looking at goals or runs, or also judging the quality of each attack? If one team takes 10 shots for 0.8 xG, and another takes 3 shots for 1.2 xG — who is actually more dangerous?

My next project will be building a season-long dataset for Chattogram's local league. It cannot be done alone — it will need a couple of volunteers, a standardized log sheet, and match-by-match verifiability. If that succeeds, we can say for the first time which Bangladeshi domestic football team is genuinely good, and which only looks good on the scoreboard.

xG Chattogram started with a single Facebook post. Those 5,200 shares told me — people are hunting for numbers, because they sense a gap in the official story. I am still doing the work of filling that gap — sitting at the ground, building tables, interrogating every number until it confesses its context.